The goal is to teach a logographic writing system to severely and moderately retarded, nonreading, adults. The theoretical interest of the research lies in the opportunity to osberve basic reading-like processes divorced from the difficulties of discriminating and learning English words. We will investigate the ability of retarded nonreaders to (1) use symbols to present objects and relations in the environment, (2) interpret a sequence of symbols as a complex event, and (3) acquire a sufficiently large vocabulary for meaningful "reading". If successful, the subjects should be able to "read" short simple stories and demonstrate understanding. If we are unsuccessful, the point of failure may suggest causes of the disability such as inability to integrate sequences or symbols, low capacity of long term memory, or the like. The specific training procedures are outlined in the body of the proposal. Verbal instruction will be supplemented by manipulation of dolls and objects to illustrate actions. This will be done by the experimenters as demonstrations and by the subjects as test responses. In this way, we hope to avoid more rote learning of paired-associates. We also plan to use Premack's methods for investigating syntactical knowledge. The results will have practical implications as well as theoretical. It is possible that learning an easy writing system first will facilitate later learning of true reading. Another possibility, if the method is successful, is development of a logographic language for mentally handicapped which could serve a function similar to sign language for the deaf.